VOL.
7,
NO.
10. O C T O R E R 1 9 6 8
Ligand-Induced pK Changes in Chymotrypsin* David M. Glickt
6.7 and the raising of the pK of a group originally with pK = 8.6. Formanilide shows an effect only on the pK = 8.6 group. Several other ligands, including L-(1-tosylamido-2-pheny1)ethyl chloromethyl ketone, were tested. These experiments are interpreted as indicating that the ammonium-carboxylate salt bridge and the imidazole-hydroxyl hydrogen bond shown to exist in the tosylchymotrypsin crystal are formed or strengthened by the complexing of a ligand.
ABSTRACT: When neutral ligands, such as anisole and coumarin, are added to an unbuffered a-chymotrypsin solution in the pH range 6-7.5, the release of a proton can be detected. In the pH range 7.5-9 the uptake of a proton is seen. The magnitude of this effect is related to ligand concentration in a way that indicates complexing with the enzyme at the substrate binding site. These data show that binding of a ligand is associated with the lowering of the pK of a group originally with pK =
C
with small changes in pK values which occur when chymotrypsin binds ligands.
hemical evidence has indicated the existence of a hydrogen-bonded serine-histidine pair at the catalytic site of chymotrypsin and an N-terminal isoleucine-aammonium-carboxylate salt bridge necessary to maintain the integrity of the enzyme's binding site (Bernhard, 1968, p 238ff). The recent X-ray crystallographic studies of tosylchymotrypsin have confirmed these inferences (Mathews et af., 1967), but still to be settled is whether these intramolecular interactions are present in the native enzyme which is not associated with a ligand. This is an important issue because the tosyl group has a substrate-like structure; indeed, tosyl chloride is an efficient and selective chymotrypsin active site reagent because of this similarity. The binding of a ligand to chymotrypsin, as indicated by the K, of ATA,I depends upon a dissociating group with a pK near 8.3 (Himoe and Hess, 1966). Because this pK is close to that for a transition in the enzyme detected by a change in specific rotation and associated with the salt bridge involving the a-ammonium group of isoleucine 16, the binding site of the enzyme is believed to be associated with this salt bridge (Oppenheimer et af.,1966). If the dissociating group can affect binding of ligands, then the binding of ligands would be expected to affect the dissociating group. Previous experiments have not been designed to examine small changes in pK values. The studies presented here deal
Materials
*This work was performed at the Biology Department, Brookhaven National Laboratory, Upton, New York. Receiaed April 15, 1968. A preliminary report of this work was presented at the 154th National Meeting of the American Chemical Society, Chicago, Ill., Sept 1967. This work was supported by the Atomic Energy Commission. t Present address: Department of Biochemistry, Marquette School of Medicine, Milwaukee, Wis. 'Abbreviations used: ATA, N-acetyl-L-tryptophanamide; ATEE, N-acetyl-L-tyrosine ethyl ester; DIP-chymotrypsin, diisopropylphosphochymotrypsin; TPCK, ~-(l-tosylamido-2pheny1)ethyl chloromethyl ketone; TPMK, (l-tosylamido-2pheny1)ethyl methyl ketone.
Three-times-crystallized a-chymotrypsin, three-timescrystallized chymotrypsinogen A, and salt-free twotimes-crystallized trypsin were purchased from Worthington Biochemical Corp.; ATEE from Mann Research Laboratories ; TPCK from Cyclo Chemical Co.; anisole from The Matheson Co., Inc.; benzenesulfonamide and formanilide from Distillation Products Industries; coumarin and a-naphthol from Aldrich Chemical Co., Inc.; DFP from Boots Pure Drug Co.; Tris from Sigma Chemical Co.; and standard NaOH solutions from Anachimia Chemicals, Ltd. A sample of purified chymotrypsin was prepared by passing a solution of commercial chymotrypsin over a Sephadex G-25 column equilibrated with 0.001 M HC1. Chymotrypsinogen free of chymotrypsin was prepared by treating commercial chymotrypsinogen at 4" with 0.001 M DFP at pH 7.2 in 0.05 M sodium phosphate buffer for 4 hr, followed by dialysis against 0.001 M HC1, and lyophilization. Because this treatment did not completely abolish chymotryptic activity, the procedure was repeated with 0.004 M DFP. The preparation then had no chymotryptic activity, but incubation with 4 x by weight trypsin within 4 hr released chymotryptic activity corresponding to 4 3 x of the DFPtreated material being chymotrypsinogen. DL-TPMK was synthesized from DL-3-amino-4phenyl-2-butanone hydrochloride (purchased from Cyclo Chemical Co.) by reaction with tosyl chloride in pyridine overnight at room temperature. The product was recrystallized from acetone and petroleum ether to a melting point of 103.5-104.5" uncor. Elemental analysis (Schwartzkopf Microanalytical Laboratory) showed the following. Anal. Calcd for CI,HIsNOaS: C, 64.33; H, 6.03. Found: C, 64.39; H, 5.96.
LIGAND-INDUCED
3391
P K CHANGES IN CHYMOTRYPSlN
BIOCHEMISTRY
/O
pMOLES CHYMOTRYPSIN
1
I IMM ANISOLE) 1: The ligand-dependent release of protons to the medium at three levels of chymotrypsin. Conditions: 0.10 M KCl, pH 6.5, and 30".
FIGURE
Methods
3392 OLJCK
Assays for chymotryptic activity were performed using a Radiometer TTTl titrator and ABUl autoburet unit, adapted to a Leeds and Northrup Speedomax G recorder. The flexible cable from the autoburet unit drove a ten-turn helipot to give, through a balanced bridge circuit, an output equivalent to one, four, or ten times the full-scale deflection of the recorder. Enzyme activity was determined from the rate at which 0.050 M NaOH was used to maintain a given pH (7.8, unless otherwise noted) when chymotrypsin (usually 1-5 pg dissolved in 50 pl of water) hydrolyzed ATEE (added in 50 pl of acetonitrile) in 4.00 ml of water containing 0.02 mmole of Tris, 0.4 mmole of KCI, and 0.08 mmole of CaClz at 30". When an inhibitor was present in the assay, it was added in 50 pl of methanol, and the same volume of methanol was added to each assay in the corresponding control series. Chymotrypsin was inactivated by TPCK at pH 7.0 at 30" in a 10-ml solution containing 10 mg of chymotrypsin, 1 mmole of Tris, and 0.2 mmole of CaClz (or 1 mmole of sodium phosphate and no CaClZ), and a varying amount of TPCK in 0.10 ml of isopropyl alcohol. At five timed intervals between 0.5 and 10-30 min, depending upon the TPCK concentration, 100-pl aliquots were removed and added to 2.00 ml of a pH 4.0 solution containing 0.1 mmole of sodium acetate. Aliquots of these diluted samples were then assayed for enzyme activity. The short-term rate of inactivation of chymotrypsin by TPCK at 30" was determined at pH 6.5 with 49.5 ml of a solution containing 250 mg of chymotrypsin, 5 mmoles of KC1, and 1 mmole of CaC12. Before addition of inhibitor, two 100-pl aliquots were taken and added to 5.00 ml of 0.01 M HCI at 0". Then 0.500 ml of 0.035 M TPCK in methanol was added to the enzyme. At 30 and at 111 sec 100-pl aliquots were removed and diluted as were the two before. The four diluted aliquots were then assayed for enzyme activity. The release of protons from chymotrypsin was measured in the titrator-buret-recorder apparatus described above. At 30" a solution containing 250 mg (except as noted) of chymotrypsin, 5 mmoles of KC1, and 1 rnmole of CaClz was adjusted to the desired pH
FIGURE 2: Graphical determination of Ki and Jmax for anisole at pH 6.5. The data from Figure 1 are plotted to give as the intercept on the abscissa, -Ki, and as the slope, l/Jmax. The triangles are for 2 pmoles, the squares for 6 pmoles, and the circles for 10 prnoles of chymotrypsin.
and to 49.75 ml. The compound being investigated was dissolved in methanol and a 100-p1 portion was added to the chymotrypsin solution. The volume of 0.050 M NaOH required to regain the original pH was recorded. The procedure was then repeated until a total of five portions had been added. At the higher pH range (8-9), autodigestion and some absorption of atmospheric COZ caused a high enough background rate of alkali addition that the uptake of protons on addition of a ligand was seen as a temporary cessation of the background addition of alkali. The linear rate of renewed alkali addition was extrapolated to that time when the ligand was added and the magnitude of the proton uptake was then calculated. Because of this complication the data for the higher pH range are not so accurate as those for the lower pH range. When the ligand tested was TPCK, the experiment was performed as above, but only one addition of TPCK was used to elicit proton release or uptake in a single experiment. The same experiment at pH 6.5 was performed on trypsin and on chymotrypsin-free chymotrypsinogen. Between experiments the electrode and reaction vessel were washed with acetone to remove traces of ligands and with 0.01 M HCl to remove traces of adsorbed protein. Results When anisole was added to chymotyrpsin at pH 6.5, protons were released to the medium. As the protons could not possibly have come from the anisole, they must have come from the enzyme. Figure 1 shows that the magnitude of the proton release, j , is directly proportional to the amount of enzyme present, and that there is a saturation of the binding site as the ligand concentration, I , is raised. These data from equilibrium states may be treated like kinetic data (Dixon and Webb,
VOL.
7,
NO.
Io,
OCTOBER
I968
I
-0.2 -
\
I
I-
I
4: J,, (the protons released per ligand bound) as a function of pH. The circles are for anisole; the triangles for coumarin; and the crosses for formanilide. The lines are calculated from the pK and pK' values in Table I. Conditions: 0.04 mM chymotrypsin, 0.10 M KC1, 0.02 M CaC12,1-5 mM anisole, 0.3-1.5 mM coumarin, and 1-10 mM formanilide at FIGURE
-1-2-3-4-5-
MM ANISOLE
y pH 8.5
30".
3: Graphical determination of J,,, at several pH values. The anisole concentration, I , per fimoleof protons released, j , is shown as a function of I for various pH values. I,,,,, is the reciprocal of the slope of each line divided by the amount of enzyme present. Conditions: 0.04 m~ chymotrypsin, 0.10 M KCl, and 0.02 M CaCh at 30". FIGURE
1958, p 21) to give a binding constant, Ki.2 In Figure 2 l / j is plotted against Z to give as the intercept on the Ki determined abscissa, -Ki, and as the slope, l/Jmsx. in this way is the same as the inhibition constant determined from inhibition of chymotryptic hydrolysis of ATEE. The proton-release Ki is 2.5 m~ and the kinetic Ki is 2.7 m, indicating that the release of protons occurs when anisole binds to the substrate binding site. The reciprocal of each slope in Figure 2 may be divided by the amount of chymotrypsin present to give Jma,.This value is constant for these three experiments: 0.067 proton/anisole bound. The proton-release experiment was performed over the pH range 6.0-9.0 and the results are shown in Figure 3. Binding of substrates becomes less tight as the pH is raised above 7.5 (Himoe and Hess, 1966) and it might be expected that binding of anisole would The protons released, j , is related to the ligand concentration, I , the dissociation constant of the enzyme-ligand complex, Ki, and the limiting value for j asZ increases, J,,,,,, by the equation, j = JmaxZ/(Ki I ) . J,,, is always given as moles of protons released per mole of enzyme present, so that it may be taken to mean protons released per ligand molecules bound. 2
+
be similarly affected. This is seen in Figure 3 by the intercepts on the abscissa occurring at -Ki for pH values up to 7.5, but above that pH the intercepts are increasingly negative. When the J,,, values, the reciprocals of the slopes, from Figure 3 are plotted as a function of pH, a complex curve is obtained (Figure 4). The release of 51 proton at constant pH is due to a change a priori in pK of a dissociating group. The digerence between the two dissociation curves characterized by pK (the original value) and pK' (the ligand induced value) is a bell-shaped proton release us. pH which curve. The height of this curve at any pH is J,., has been experimentally determined. The height of this difference curve at its maximum and the value of the pH at which this maximum occurs can be used to calculate pK and pK'. The same experiments were performed on other inhibitors. The results are given for coumarin and formanilide in Figure 4. The former case resembles the experiments with anisole, but the latter case shows no proton release near neutrality, only a proton uptake in the higher pH range.4 With both these compounds the Ki values determined from proton-release data (0.8 and 3.8 mM, respectively) agreed with the kinetic Ki values. 3 In general, when b is the fraction of the dissociating group in the unprotonated form and b' is the fraction of the perturbed log dissociating group in the upprotonated form, pH = p K (b/(l - b)) and pH = pK' log (b'/(l - b')) and their sum is 2pH = pK pK' log (b/(l - b ) ) log (b'/(l - b')). At that p H where J is largest, pH = (pK pK')/2. Therefore, at that pH, (1 - b)/b = b'/(l - b'). As J is defined as b' - b, b = (1 - J)/2 and 6' = (1 J)/2 and pK = pH log ( 1 J)/(1 - J ) and pK' = pH log (1 - J)/(l J ) . ' I t may be noted, however, that formanilide first added to chymotrypsin interfered with the release of protons subsequently elicited by anisole at pH 7.
+
+
+
LIGAND-INDUCED
+
+
+
+
+
IJK
+
CHANGES
+
I N
+
3393
CHYMOIKYPSIN
BIOCHEMISTRY
I
0
25
50
75
100
S1C
5 : Rate of short-term inactivation of chymotrypsin by TPCK. Conditions: 0.04 m chymotrypsin, 0.10 M KCl, 0.02 M CaC12,and 0.35 m TPCK, at pH 6.5, 30". FIGURE
a-Naphthol, with a kinetic and proton-release Ki of 2.6 mM also elicited a proton release at pH 6.5, but, because of the acidity of the phenolic hydroxyl group, it was not investigated further. Similarly, benzenesulfonamide, with a kinetic Ki of 5.9 mM, showed a J,,, at pH 6.5 of 0.155, but the acidity of the sulfonamide group prevented a more extensive investigation. It was desired to test TPCK (Schoellman and Shaw, 1963) in the same manner as anisole, coumarin, and formanilide. As the proton release (or uptake) was determined at only one concentration of TPCK at a given pH, it was necessary to know the K,5 for this compound. The pseudo-first-order rate constant for the inactivation of the enzyme by TPCK was determined as a function of TPCK concentration. The insolubility of TPCK precluded measurements above 0.35 mM. (Substitution of phosphate buffer for Tris and CaClz made no difference in the results.) The K, determined by conventional graphical means was 0.3 mM. Because in these experiments TPCK was not in large excess over chymotrypsin, this estimated K, was used as a dissociation constant to calculate the free TPCK concentration initially present in the chymotrypsin-TPCK solution. Incorporating this small correction (less than l o x ) , into a recalculation of K, produced a corrected value of 0.27 mM. To confirm that the release of protons was associated with formation of a reversible enzyme-inhibitor complex and not with the following alkylation step in which HCl would be formed, the activity of chymotrypsin was determined before and immediately after addition of TPCK. The results, shown in Figure 5, indicate that in the few seconds following TPCK addition there was no sudden permanent loss of activity. During this time there was a release of protons which, if due to alkylation, would correspond to about 3 0 x inactivation. The data shown in Figure 6, then, reflect changes in pK of groups on the enzyme. Because of the acidity of the sulfonamide K , here means (k2
RCI
3394 GLICK
+ k3)/kI in the sequence
ki ka + EzH e [RCI---EzH] +EzK+HCI kz
RCl is TPCK, EzR is the irreversibly inactivated enzyme, and the intervening species is the reversibly bound enzyme-TPCK complex.
FIGURE 6: J,,, (the protons released per TPCK bound) as a function of pH. The line is calculated from the pK and pK' values in Table I. Conditions: 0.04 mM chymotrypsin, 0.10 M KC1, 0.02 M CaC12,and 0.3 mM TPCK at 30".
hydrogen of TPCK (KCzdy et al., 1967), it was not possible to extend the data to a higher pH, although it is evident that a group with pK above 8 is being affected by TPCK binding.6 Neither trypsin nor chymotrypsinogen gave any release of protons at pH 6.5 on addition of TPCK. Because the chymotrypsin-free chymotrypsinogen contained DIP-chymotrypsin, it was also seen that this DFP-inactivated material also showed no proton release. To assure that proton release was due to chymotrypsin itself and not any low molecular weight contaminant in the commercial preparation, a sample of chymotrypsin purified by gel filtration was tested with cynaphthol and with TPCK, giving results indistinguishable from those obtained with the commercial sample. The pK and pK' values are presented in Table I. No pK' with TPCK for the pK = 8.6 group is shown only because it was not quantitatively evaluated, not because there was no shift of pK. Discussion Bender described the uptake of a proton by chymotrypsin when it binds benzyl alcohol above pH 7.5 (Bender and Wedler, 1967), but his data show, not a bell-shaped curve expected for a shift in pK, but a sigmoid curve that mysteriously indicates the complete abolition of an acidic group with pK = 8.8. A possible although not completely satisfying explanation is that there is a pK shift of more than 2 pH units, and, had it been possible to continue the experiment above pH 9.5, a bell-shaped curve would, in fact, have been found. The magnitudes of the pK shifts reported here (Table -
~
~
...
. ~ .~ -. _ .
For purposes of comparison with TPCK, it was of interest to perform the proton-release experiment on the nonhalogenated analog, TPMK. When DL-TPMKwas synthesized and examined, no etfect could be demonstrated up to a concentration of 0.7 mM. The insolubility of TPMK prevented a closer approach to the Ki, kinetically determined to be 2 mM, which rendered this particular experiment inconclusive. 6
VOL.
7,
NO.
10,
OCTOBER
1968
I) are not obviously related to structures of the compounds, but the results unequivocally indicate that binding of a ligand to the active site of chymotrypsin can change the pK of groups on the enzyme. Any identification of the groups whose pK values are being perturbed must be tentative, but it is tempting to identify the pK = 6.7 group as the imidazole of histidine 57 which has been shown to have pK = 6.7 and is part of the active site (the evidence is reviewed in Baker, 1967, p 130ff). The a-ammonium of isoleucine 16 might be the pK = 8.6 group because of its demonstrated interrelation with binding properties (Oppenheimer et al., 1966). X-Ray crystallography of a chymotrypsin derivative in which the active site serine is tosylated shows a salt bridge between the a-ammonium group of isoleucine 16 and the P-carboxylate group of aspartic acid 194, and a hydrogen bond between the imidazole of histidine 57 and the tosylated oxygen of serine 195 (Mathews et al., 1967). The special reactivity of serine 195 as a nucleophile has been ascribed to this interaction (Cunningham, 1957). If the tosylchymotrypsin crystal is a reliable guide to the intermolecular interactions of the ligandbound molecule in solution, we can conclude that the salt bridge and the hydrogen bond exist in a chymotrypsin-anisole (or other ligand) complex in solution. This paper compares the ligand-bound state, in which the salt bridge and hydrogen bond exist, with the unbound state. It is suggested that when chymotrypsin binds a ligand, it is histidine 57 that becomes a stronger base and isoleucine 16 that becomes a stronger acid. This is consistent with the formation of the salt bridge and hydrogen bond accompanying the binding of a ligand. The free enzyme does not possess these interactions, the ligand-bound form does, Bringing together an ammonium and a carboxylate group in the interior of a protein molecule is energetically favorable. Deprotonation of the ammonium group disrupts this favorable arrangement. Therefore the ammonium group as part of this pair is a stronger acid than it is alone. The carboxylate group, in the same way, becomes a stronger base; proton-release studies at a lower pH range should detect this. A similar argument applied to the histidine-serine interaction indicates that in the hydrogen-bonded pair histidine is a stronger base than it is alone. Possibly the strengthening of these interactions is achieved by complexing of the ligand with one of several conformers of the free enzyme which exist in equilibrium. This one form possesses the two intermolecular interactions and the binding site for the ligand. As the ligand binds to this conformer, the equilibrium among all the conformers is displaced in its favor. A similar explanation was offered (Hummel and Witzel, 1966) to explain the pHdependent uptake and release of protons when ribonuclease binds nucleotides.7 According to this interpretation of the data the effect of formanilide upon 7 This analysis follows logically from the studies of Lumry (Parker and Lumry, 1963; Yapel er al., 1966; Lumry and Kim, 1967) on the various conforiners of chymotrypsin and their varying ability to bind ligands.
TABLE I :
Values of pK and pK' Measured for Various
Ligands. Inhibitor Anisole Coumarin Formanilide TPCK
pK
pK'
pK
pK'
6.7
6.5 6.55
8.6 8.5 8.7 b
8.9 8.8 9.3 b
6.65 a
a
6.7
6.1
a No shift in pK was detected in this pH region for formanilide. b There was a positive shift in the pK value in this pH range for TPCK, but because of the acidity of the sulfonamide hydrogen, the magnitude of this shift was indeterminate.
only one of the dissociating groups may mean there can be a form of chymotrypsin that has a binding site and salt bridge but not the hydrogen bond. It is, of course, not only incidental that the bound form has the histidine-serine interaction that makes serine an effective nucleophile. It is this interaction, triggered by the binding step, that is essential for the next step in catalysis, the nucleophilic attack of the P-oxygen of serine on the carbonyl carbon of a substrate. Acknowledgments Thanks are due to Drs. C. H. W. Hirs and D. E. Koshland, Jr., for the hospitality of their facilities at Brookhaven National Laboratory during the author's tenure of a Brookhaven Fellowship, and to Dr. R. Lumry for making available a draft of a chapter to appear in a future publication. References Baker, B. R. (1967), Design of Active-Site-Directed Irreversible Enzyme Inhibitors, New York, N. Y . , Wiley. Bender, M. L., and Wedler, F. J., Jr. (1967), J. Am. Chem. SOC.89,3052. Bernhard, S. (1968), The Structure and Function of Enzymes, New York, N. Y., Benjamin. Cunningham, L. W. (1957), Science 125,1145. Dixon, M., and Webb, E. C. (1958), Enyzmes, New York, N. Y., Academic. Himoe, A., and Hess, G . P. (1966), Biochem. Bioplzys. Res. Commun. 23,234. Hummel, J. P., and Witzel, H. (1966), J. BioL Chem. 241,1023. Kkzdy, F. J., Thompson, A., and Bender, M. L. (1967), J. Am. Chem. SOC.89,1004. Lumry, R., and Kim, Y . (1967), 154th National Meeting of American Chemical Society, Chicago, Ill., Abstract 194. Mathews, B. W., Sigler, P. B., Henderson, R., and Blow, D. M. (1967), Nature 214,652.
LItiAND-INDUCED
PK CHANCILS I N
3395
CHYMOTKYPSIN
I4 I O C H E M 1 S'I' K Y
Oppenheimer, H. L., Labouesse, B., and Hess, G. P. (1966), J. Biol. Chem. 241,2720. Parker, H., and Lumry, R. (1963), J . Am. Chem. SOC. 85,483.
Schoellmdn, G., and Shaw, E. (1963), Biochemistry 2, 252. Yapel, A., Lumry, R., and Han, M. H. (1966), Federation Proc. 25,650.
Some Physical and Chemical Studies on Two Polypeptide Components of High-Density Lipoproteins of Human Serum" V. Shore and B. Shore
Two different polypeptides separated from the protein moiety of high-density lipoproteins of human serum were found to be similar in molecular weight but very different in amino acid composition. One of the peptides, characterized by carboxyl-terminal glutamine, contains no histidine, arginine, tryptophan, or cysteine. Its amino acid composition is: Lysis,Cy~~,Asp,,Thr~l,Ser~~, Glu26,Pro?,GlY6, Alas, VaLo, Metz, Ile2,Leu14,Tyre,Phe7; the total number of residues is 133 and the molecular weight from amino acid composition is 14,900. Sedimentation equilibrium experiments yielded molecular weight values of 14,300 and 14,900 for the polypeptide R-Gln in urea solutions and in guanidine hydrochloride solutions, respectively. The other polypeptide, characterized by carboxylterminal threonine, contains no isoleucine, cystine, or ABSTRACT:
T
he protein moiety of high-density lipoproteins (1.065-1.195 g/cc) of human serum is heterogeneous (Shore and Shore, 1968a,b). Polypeptides with carboxylterminal glutamine and threonine were found in comparable amounts by hydrazinolysis and by carboxypeptidase experiments on the lipid-free proteins from the 1.126-1.195-g/cc (HDL,) and the 1.083-1.124-g/cc (HDL2)lipoprotein fractions. The present report describes some physical properties, obtained principally by the method of equilibrium sedimentation, and the amino acid compositions of these two polypeptide subunits which were separated by DEAE-cellulose chromatography in 8 M urea. Experimental Section Materials. The lipoproteins of density 1.126-1.195 g/cc (HDLp), which were used as the source of protein in these experiments, were isolated from human serum
3396
* F r o m the Division of Biology and Medicine, Lawrence Radiation Laboratory, University of California, Livermore, California 94550. Receimd June 17, 1968. This work was performed under the auspices of the U. S. Atomic Energy Commission.
SHORE ANV S H O R t
cysteine. Its amino acid composition is: Lyslo,Hisr,Args,Asp13,Thr,,Ser~,Gluz4,Pros, Gly,, Alalo,Val,, Metz,Leuzo, Tyr4,Phea,Trp4;the total number of residues is 133 and the molecular weight from amino acid composition is 15,500. Sedimentation equilibrium experiments on the polypeptide R-Thr in guanidine hydrochloride solutions indicated homogeneity with respect to molecular weight; however, the molecular weight value 31,400 indicates that the polypeptide R-Thr exists as a dimer in guanidine hydrochloride solutions. Sedimentation equilibrium experiments on the polypeptide R-Thr in urea solutions and in dilute salt solutions containing sodium dodecyl sulfate indicated heterogeneity with respect to molecular weight and weightaverage molecular weights of approximately 20,000.
by ultracentrifugation as described previously (Shore and Shore, 1967). The protein moiety was obtained with essentially complete recovery in lipid-free (
